Hi, my post is focusing specifically on YouTube since I observed the following categories have less intrusive solutions or privacy focused solutions, even if they are paid:
Operating Systems (Linux, for example)
Instant Messaging (Element, for example)
Community Messaging (Revolt, for example)
E-Mail (Proton, for example)
Office (libreoffice, for example)
Password Managers (Bitwarden, for example)
However, how do we distribute videos and watch them without data collection? I am NOT asking how do I use a privacy-focused front-end for YouTube, by the way, I am aware they exist.
I am wondering how we obtain a FOSS solution to something super critical such as YouTube. It is critical since it contains a lot of educational content (I'd wager more than any other platform), and arguably the most informative platform, despite having to filter through a lot of trash. During COVID, we even saw lecturers from universities upload their content on YouTube and telling students to watch those lectures. (I have first-hand experience with this at a respectable university).
I refuse to accept that there is nothing we can do about it.
I refuse to accept that there is nothing we can do about it.
I don't think you quite understand just how stupendous the amount of data Google processes from YouTube alone is. There is basically no way for hobbyists to provide an equivalent service. Very few companies have those kinds of resources. If you want, you can of course try running a PeerTube instance, but you rather quickly run in to problems with scaling.
I find it almost miraculous YouTube exists to begin with. It is no accident Google has very few competitors on that front, and I don't think YouTube is even profitable for them. Without Google's deep pockets and interest in monopolizing the market, YouTube would have withered a long time ago.
Trust me, I want a solution too. But 500 hours of content are uploaded to YouTube every minute. All of that is processed, re-encoded, and saved with multiple bitrates. You can't compete with that. YouTube might eventually keel over from Enshittification and its own impossibility, but replacing it with anything meaningful will be a challenge.
[...] I don't think YouTube is even profitable for them.
Correct. Even Google, one of the richest companies in the world, is struggling to afford the massive infrastructure required to run YouTube. That's why they've been cracking down on ad-blocking software lately.
Also, this is likely why they've been pushing their new updated Chromium-based infrastructure for web browsers, which will prevent ad-blockers from working on websites. If you're not using Firefox or Safari to browse the Internet by now, you should switch. They're the only independent browsers not using the Chromium framework.
Restaurants don't take steaks off the menu because they aren't are profitable as salads. One date wants a salad, the other wants steak, they make less profit on the steak plate, but the average of the two is profit enough.
It's ridiculous to look at any one service of these behemoth monopolies as an island - They are one collective thought, EVERY SINGLE PIECE does not have to be to enshittified to generate the biggest possible profit.
While I do agree with you, I also see twitch, TikTok and Patreon presenting models that are quite competitive with YouTube.
From a privacy perspective, free junk content like TikTok, YouTube and twitch will always be hard coupled with targeted advertising.
But Patreon (and onlyfans for that matter) do offer a model that can work without ads.
In fact, if Patreon also introduced an ad-supported tier and allowed you to more broadly see other content aside from the direct person you sponsor, it could probably grow quite a lot.
Counter-point : every single one of the videos uploaded to youtube already lives on the creators hard drive, usually in a much larger format. All that's needed is for them to create torrents for them.
I think the largest challenge though is maintaining the distribution and managing the associated upfront costs.
Existing large content producers could likely afford to handle this but new producers could struggle paying to seed their content.
Though I do think overall this is more achievable than people give it credit for:
YT videos don't need huge bandwidth for a sustained period; only for short bursts. Most views come in within a week.
Content is probably localized to specific countries. Less need to replicate across the globe.
Let the source prefer to seed the highest quality and other peers downsample and replicate as needed.
Doesn't need YT scale. Tons of YT "content" is spammers leeching essentially free hosting from YT. No one needs to seed their videos if they don't want to.
1080p is still fine for YT videos. h265 is very efficient (though downsampling 265 isn't great). Don't need 4k for most videos.
I'd have agreed but hundreds of fmovies and similar sites exist on the high seas that provide free streaming of millions of HD content (movies, web series, etc.) somehow. They use some third-party video host that is magically able to concurrently serve millions of people.
Maybe the solution to YouTube is something similar to BitTorrent. It would make more sense for the protocol to preload the first chunk and to use a codec that can start with a lower res image and then fill in the resolution in subsequent passes. And on the front end, something like Lemmy would work, where channels and posts can be federated.
Considering the number of people who have 1gps symmetric bandwidth today, such a system should be able to technically work.
now I will admit that peertube is lacking content, but when you make something put it there. When you want something search there first and check out youtube last. This rewards those who publish there with your eyeballs
I'm not sure if you can replace YouTube. It's too popular and has been a mainstay of the Internet for 19 years. We won't be able to convince people to just up and leave YouTube.
Best case scenario is to lead by example and start sharing videos from PeerTube.
The technology mostly exists. The most important question is always how do you get people to use it.
The only way I see people using decentralized solutions is by having one interface where you can watch decentralized content as well as YouTube. That way they don't loose any of the content or convenience.
No one ever bothers to open up two apps for videos, that is why a single app solution is the only way.
The unique selling point of decentralized video plattforms atm is 1) you can watch what is banned on YouTube 2) you are not beholden to the YouTube algorithm for conent.
So if we can sell that to users and not have them loose any convenience or UX, you can slowly start replacing YouTube.
Monetization is also an important point, but others have addressed this.
If they integrate some self-hosted analytics and monetization mechanics, like Matomo and Stripe, to it, then it'll be a feasible alternative to YouTube.
Ok so first let’s go over what YouTube provides:
Storage, community tools, search algorithm, add sense, authority over copyright, front end.
Realistically you could probably cover the front end, search algorithm, and community tools with FOSS collaboration.
Everything else gets harder.
For storage, the VAST swaths of data, and forever growing nature of YouTube storage nearly guarantee its market dominance alone… if they can contain that infinitely growing monster forever. Its their greatest strength and can also be its Achilles heel. I would propose that video hosting would be covered by the creatives. This change creates a ripple effect that effect all the other challenges, but immediately raises the bar for entry, and with the exception of the highest earning creators, videos would have to be cycled out when their earning capability falls below cost to host. But! This has good sides, like the best videos would linger and bad videos would fall off increasing the quality of what remains. Creatives would have more control over their videos. You could also have a system that rotates videos between a cold storage and live videos, where cold storage would use a torrent like system vs the streaming of a live system, which would allow cheap storage of low earning videos to still have them available for those who could wait.
Copyright, so with the creatives holding the keys to the content, this new youtube would only facilitate the connection and front end, but would not regulate it. So copyright claims would have to be handled by the creatives. This is a sharp as hell double edged sword! You won’t be copyright trolled as successfully any more BUT your odds of ending up in court could be higher as there is no way to appease the record labels and what have you so readily.
There would also not be a method to scan the videos to easily find other people who are stealing YOUR content either. And you would have to deal with the person stealing your content directly.
And ad sense. Without a unifying front to bargain with advertisers, it will be like the Wild West. Most advertisers don’t have assurances of enforced standards and will be very timid to employ this new system. They would all have to vett creatives separately, and it would work allot like Sponcers do now, but ultimately i think it would be a boon, but for a wile the money won’t be there.
So i put more thought into this… assuming this was how a youtube competitor turned out. The negatives would begin to force certain human behaviors to mitigate risk. You would see guilds/channels form. This covers the weakness of the Wild West. Groups can bargain with more leverage from sponcers and demand more money in exchange for more consistency, these guilds/channels can also hire a lawyer on retainer if large enough to handle litigious tasks, and advise its members though copyright dangers. If it when it goes to court they can handle hiring of additional representation. The guild/channel would have say as to who they admit to the group, so they can expel risky members. But like joining an HOA creatives will have to adhere to the channels rules. But without a monolith controlling everything, you could find a guild/channel that has terms you agree with. This would bring a lot of the status quo youtube brings, but with everyone’s goals more aligned
For the algorithm,i would recommend using a hash tag system (i know they are not called hash tags but I’m in a stream of consciousness here) give creators the freedom to label hashtags to their content. Though to avoid gaming them, the value of views/upvotes is divided equally amongst all the tags, so if you put #hollow_knight as your only tag, you get more weight on a smaller net. Or if you act like an Amazon reseller and dump every single hash tag on you video to throw the widest net, you get a more shallow weight in each tag. I would count views AND like for this. Likes would be weighted more due to needing engagement. I probably would recommend not having down votes weighted either way, but obviously shown. And subscribing just guarantees the viewer gets notified at the top of the page.
The likes of popular youtubers with good content like Tom Scott and GamersNexus do not even make the list at all.
Good channels like Stories to Old that aren't big, but well produced probably won't be able to make it at all with this setup unless they form a coalition with other small creators to pay for hosting costs and have someone with the expertise to manage it. That cost would severely cut into what they would be able to live off of.
The most likely scenario is the platform becomes a wasteland of clickbait and child-friendly clickbait because that is what gets the most watch time.
So what YouTube is now. But there will be a higher bar for entry. I said as much. I fully expect groups to form and would welcome them. And the hash tag system would allow greater means of finding content that people want to actually watch, and still allowing these content farms to operate.
But this is a discussion about possible YouTube replacement, and realistically i don’t see another company that could handle the infinite demands of free on demand video streaming that we would have been as our new masters. I took inspiration from the Fediverse in this regard. The FOSS collaboration may be able to stream line the hows and specifications expected to have creatives connect their content to the collective.
Honestly the biggest thing all of us is missing to take it down is financial capital.
To get the kind of capital you need to take down YouTube, you need investment money from the kind of investors who will force you to enshittify to afford paying them back.
The financial issue is the biggest one, when it comes to any and all of these.
There are two YouTubes. One is the "creator" YouTube, algorithms, numbers blah blah
The other is the actual content creator YouTube. These are the channels that people actually follow. If captain disillusion set up his own RSS feed for videos, and I had the method to subscribe to it, I'd no longer need YouTube
The argument that YouTube has the algorithm and recommendations etc is moot, that's the same job that every network does, you could absolutely replace this
The video content would have to be self hosted probably. How it used to be. So we need all these tools to eat YouTube's lunch
Hosting video is really expensive. Making video is really expensive. YouTube was losing money for about 15 years despite having a monopoly on online video for most of that time and the best advertising tech in the world. I don't think it's possible to make a free competitor to YouTube.
On the paid side, there's plenty of streaming services that are making money. But you have to be already established in order to get a contract. And since you will typically have to use social media in order to get past that initial barrier, it might as well include YouTube.
However, my guess is that YouTube makes the majority of it's money from larger channels. If the larger channels all join paid streaming services(e.g. Nebula) then gradually that may be able to bring YouTube down.
I was just reading this issue on Github last night and I really don't see how PeerTube is any better than a traditional server for hosting videos. The peer part of it seems to have such a miniscule impact on the whole thing that it just feels like a gimmick. I've read that the biggest problem for PeerTube instance hosts is storage and not the bandwidth. The only thing that peers can save you is tiny bit of bandwidth from what I understand.
So from what I've gathered, relying on peers only for hosting the video is completely unviable. And that makes sense, especially for old, unpopular videos, there will be no peers to begin with. Even if every video on the site is being "seeded" by viewers, the reliability of connection and bandwidth would be very bad because you can't know if the peer is some guy on the dial up connection. Even in the perfect scenario where everyone had very reliable connection and good bandwidth, the fact that browsers don't support p2p protocol and rely on a hack/workaround to use it, will mean that there will be delays. So starting the video and rewinding would be painfully slow.
Is there something that I'm missing, or is PeerTube really not that much better than a "normal" video hosting server?
Peertube uses webtorrents, not regular torrents, and doesn't even hook into the larger torrent network, which is seeding most of media on the net.
You're correc, the peer part of peertube is mainly a gimmick at this point, and it's nowhere close to being what torrents already are, a decentralized hosting network.
The big problem is there are a lot of good creators who are only able to be good creators in large part because of the YouTube ad revenue they get. They would otherwise have to work normal jobs and not be able to devote the time or resources to their videos. I have little faith that enough viewers would actually pay enough money to offset the ad revenue that supports many creators. Without a way to realistically replace that financial stream there is a large chunk of YouTube that can’t migrate. Of course, that’s no loss with some of the content mills churning out crap to try and cash in on the revenue, but I’ve seen plenty of good stuff that I’m not sure would exist another way.
This seems like one of the few problems where crypto might actually be useful. It would allow people to automatically and anonymously pay both the creator and the host of that video. Maybe make it a federated system and every host gets paid based on how many Bytes they send. The creator gets a share of that money and the whole system uses something like Monero or whatever. Not sure what the costs of that would be, but I assume its not too outrageous. If it was, YouTube wouldn't be able to exist.
Basically, but I'm not sure how well it'll work longterm due to the website not really contributing anything to the system afaik. Though I have to admit I haven't looked that far into it, just posting my notreallyeducated guess.
https://lbry.com/faq/host-content
That's true, you'd definitely have to charge more than what YouTube makes with ads. But I don't think Google would keep YouTube alive if it generated only like, 10% of the money it costs them to operate.
Edit: That's why I said "it's probably not too outrageous", I know that YouTube probably operates at a loss, but I don't think the cost is so great that noone would pay to fund a service like that. Though I'm obviously just guessing, I might be totally wrong
People are working on this for general decentralized storage, some of them have existed and been functional for 5+ years, I'm not familiar with all the names but there's jstor (jstore?), filecoin, etc. When you have a system where you need to manage a database (and everybody's copy of the database is the same) but you need to do it in a decentralized, P2P way, blockchain is really the only solution. A system which records who is hosting what and allows people to buy & sell storage is exactly this: a database with some buy/sell frontend.
As nice as an idea as it is, it will never be feesible for one reason: buy in. You would have to get everyone on youtube to migrate to the same platform. Just about everyone who uses windows has gripes about it, but the masses don't migrate to Linux. Because it is change at all, and there are too many choices. I like anyone else here, would love for folks to even consider an alternative, it's a losing battle against human nature.
Network Effect is the biggest hurdle for sure. I think it it true for so many other services too. I think we can agree there is no real technical problem to solve, we only look at the technical problems because trying to "fix" the social and political issues is a lot harder. Digital Markets Act is supposed to address this but time will tell if it has any lasting impact (in the EU).
Easy solution: host an FTP with all the videos. It has existed long before YouTube was a thing.
More advanced solution: Torrent ala Pirate Bay. High quality videos have been distributed this way long before YouTube even supported 1080p. Peertube is based on similar solution as this.
The main problem is to attract content creators to the platform. The problem isn’t technical.
That isn't a solution, because YouTube provides discoverability as one of its primary draws. The primary draw for content creators is the promise of being paid via the ad network. So FTP doesn't offer 1/10th of what people go to YouTube FOR. EVEN if your theoretical FTP server had literally every piece of content in the world; people would still go to YouTube.
Recommendation systems are well studied. I don’t think it’s unreasonable to add some form of recommendation layer separate from (or on top of) the content delivery. It doesn’t need to be up to par with YouTube’s, at least before there’s any major content.
Most YouTubers rely on sponsors or Patreon. Podcasters are doing the same - many of which are self hosting. So I don’t think an ad delivery system is that needed.
I don’t see how it would have to work much differently compared to how Pocketcast or Overcast already works.
The first problem is getting content to the platform.
The key problem that needs to be solved is the monetization problem. Nostr has a potential solution though. Over the last two months alone, their users have "zapped" (tipped/donated) other users around 950K (nearly 1 mil!) USD worth via lightning and that number continues to grow. And it doesn't just make it easy to pay content creators, but to also put a portion of your "zaps" towards the relay you use or development of the software if you want. If you have a nostr account, you can easily tie it to a lightning address to send/receive tips, nostr doesn't take a fee. Relays can also portion out a bit of their zaps for the people who publish the most engaging content on their relay. The possibilities are quite extensive. And because it's over lightning, zaps happen instantly and for pennies or less in fees. Though, you can use nostr without zaps at all.
For those unfamiliar with nostr, it's a decentralized social media software much like ActivityPub/mastodon, the main use right now is as a twitter/instagram clone but there's also a reddit-style section being built up as well. Video hosting itself could be done by relays or through a P2P system similar to IPFS. Moderation abilities from the perspective of the instance/relay are identical to activitypub/mastodon. But one bonus if that if your relay goes down, you don't lose your identity, since your identity and relay are separate. And if you change apps or relays (you are typically connected to multiple relays), all your content moves with you seamlessly. And the payment/zap infrastructure is all decentralized, relays don't ever custody or manage the payments. If you tip a content creator, it goes directly from you to them. The lightning network has basically limitless transaction capacity. If you have cash app, it supports lightning, so you can already send zaps (you will need different apps to receive zaps though because cash app doesn't support the LNURL standard). Strike natively supports it. And because it's lightning, it works in every country automatically.
Long-term, if I am a content creator, which "fedi"-type system is going to be attractive to me? One where users can send me tips and mircopayments or one where they can't? This is why I think nostr is going to win out long-term over AP/Mastodon. Mastodon could add this kind of functionality but I don't get the impression they're open to it. People may not want to commit to yet another $5/month subscription to a YouTuber's patreon or nebula or whatever, but they are happy to tip 1-10c after watching a video. So there's a psychological beauty to micropayments as well. As some random person I have made like 7c on tips this month, but I've also given out plenty to other people.
Unfortunately this financing requires a populace widely adopting cryptocurrency...making it a pipe dream for mainstream use.
Tips are generally a bad model as well, which creates an incentive for rapid and pandering work (like ad supported content).
Patreon had frankly built all of YouTube that is worth watching. I think a simple payment system using real banks can be integrated into smaller hosting services.
It's all academic though, YouTube is unrivaled in ad revenue and helping you expand an audience.
25% of Americans own crypto, usage continues to grow year after year both domestically and internationally. Most people have a crypto-capable wallet on their phone (CashApp, Venmo, Paypal). It solves problems traditional financial systems can't solve well. That's a trend that has been happening for 15 years. You can be mad at it, but it doesn't change that it's true.
The biggest issue I've always heard people say when it comes to replacing a video hosting service like YouTube is needing storage space and bandwidth.
I feel like ipfs, the interplanetary file system, could be leveraged to do this but it would require a concerted effort to make a fast, stable, reliable, and federated YouTube replacement, and I imagine that we would need people to financially support it.
You get me $10B annually or so, and then we can start to talk. Your single-fiber line and homelab will handle, what, 25 simultaneous users? Just have to scale that to a billion daily users or so, no bigger.
Also yt is "super critical"? Super critical is power for ICU wards and stuff, nobody is going to have a heart failure because they can't get their daily dose of #shorts. Also gestures at Wikipedia, who is glaring at you.
I think you're giving yt way, way too much credit, but simultaneously thinking that any one of us has the financial capability to not only have but risk that kind of cash. Companies have tried and failed. Users aren't doing it, chief.
Everything seems critical when you haven't tried living without. Meat eaters can't comprehend living without meat. Car drivers can't imagine living without cars.
I wondered how people pass their time without phones. Then my autistic son started demanding holding onto my phone for every waking minute he is not at school. Now I spend my day without the phone.
Now that YouTube has stopped working on NewPipe, I've stopped watching it......and it felt a bit uncomfortable to miss my videos before bed, but now it's not a big deal. None of these things are critical. There's a near infinite world of choices available to us now. We just need to pick something else.
Well said. I have found challenging myself to limit or go without certain things has had a great impact on my happiness and contentment. Once you realise you can get on fine without one thing, it puts everything else into perspective. Similar to you, I switched phone last month and purposely didn't import any of my YouTube subscriptions to see how I'd go if I just didn't have that constant stream of interesting videos there demanding my time. I went from an hour or two of daily viewing to nothing very quickly and the only impact it had was to free up more time in my day. I used to check daily for new uploads from my favourite YouTubers and now it doesn't even enter my mind, I couldn't care less.
You get me $10B annually or so, and then we can start to talk. Your single-fiber line and homelab will handle, what, 25 simultaneous users? Just have to scale that to a billion daily users or so, no bigger.
As a PeerTube instance owner, I would say that not everyone needs to join a single instance (that would be the biggest mistake). Instead, if you can self-host one and invite people you like and know, they can provide quality content. Also, having multiple smaller instances makes it easier to moderate and have quality control. Federation and direct subscription to channels also improve instance discovery.
you offer content creators a better revenue share to make content for the new service while offering the same level of stability. there's a reason why nobody has done it.
Look at the strangler pattern in microswrvice architecture. Applying this to your scenario, set up a front end to YouTube, cache the results locally (probably host in a place that allows it). Also host videos from other platforms like peertube. Once you have a lot of users, slowly prioritize "free" videos over YT content.
It's not likely to happen, but it's the pattern that FB uses to present news. First they showed a link to the story and you'd click through, then they required more of the story, then when all were hooked, they demanded the whole story to be displayed, effectively stealing all the users and the ability to advertise.
Also it’s worth mentioning the “how to distribute content among peers” problem has mostly been solved and has for over a decade, just that nobody has built out the UX for it for a YouTube clone. Torrents exist, #freenet and #hyphanet exist, #ipfs exists, these are all excellent platforms for storing and distributing content without relying on expensive, centralized hosting. Instead, users share the burden of hosting. There’s a whole category of software that solves this problem in different ways (P2P). Unfortunately, every new generation of developers seems to want to re-invent the wheel instead of using time-tested tech that already exists but just needs a UX refresh or maybe some protocol improvements.
If you have a tube site and it says “to skip ads, install IPFS”, everybody would be using IPFS.
If you're a creator, upload to Peertube and Youtube, and promote Peertube on your Youtube channel. It's a compromise, but it's the only realistic way to pull viewers over if you're not already a popular creator. Also provide some incentives to use Peertube instead of Youtube, like early uploads.
If you're a viewer, use Peertube; and when you need to use Youtube, use a 3rd party client like pipe-viewer. Don't support ad culture, donate to creators you like instead.
Oh I understand it, and I also understand that laws can be wrong and corrupt, and shouldn't always be followed. If you think how law-abiding a corporation is is more important than protecting privacy of activists, maybe that shows your true colors.
for 3d printed gun people (not personally one of them, just browsed their subreddit once), they use some vaguely blockchain crypto related p2p video host called LBRY, not sure if that model is scalable though, as it seems to be based around free p2p hosting like torrents, although there was some mention of hosting fees, presumably in crypto? not sure
I don't have solution for videos, but I am moving back to podcasts and rss as much as possible. I want to be ready when they finally forbbid watching without ads.
But I must admit content creators are not helping, content for most of them become just job to be done with. I am aware it is not their fault and that yt is pushing them, but content is geting worse.
It is hard to compete with platform that is loosing so much money. They will also buy anyone who tries. Maybe if we start being satisfied with one resolution and quality, but that will never happen.
how do we distribute videos and watch them without data collection?
So opinion answer to the latter. Opinion answer. Don't ignore YouTube.
Steam didn't ignore Win32 and ask 10k devs to port to Linux. They partnered up with CodeWeavers, WINE and others to create Proton and it made the former task largely unnecessary.
Expand federated video services to cache all videos they stream in case the original gets dunked on. And then at the same time grow the platform.
A subsection of FOSS hates wealth, but people need to be able to lift themselves out of poverty, there has to be a profit motive and that profit has to largely go to the content creators.
Without motives and incentives you can build the most beautiful codebase ever and it won't take off.
Mass censorship is coming, so platforms that don't censor and host in countries where this is legally protected will have the advantage of growing new mega sites.